Caldesmon exhibits a clustered distribution along individual chicken gizzard native thin filaments.

Our earlier immuno-gold electron microscopic study indicated that the distribution of caldesmon (CaD) on actin filaments is not uniform and is restricted to the vicinity of the myosin filaments (Mabuchi K, Li Y, Tao T, Wang CLA (1996) J Muscle Res Cell Motil 17: 243). This suggested that CaD could effectively inhibit muscle contraction, if those actin filaments in the vicinity of myosin filaments were saturated with CaD. In the present study we further examined the distribution of CaD along isolated, crude and purified native thin filaments (NTF). Individual CaD molecules on purified NTF were visualized with the aid of a chemical crosslinker, 5,5'-dithiobis(2-nitrobenzoic acid), which efficiently crosslinks CaD to actin (Graceffa P, Adam LP, Lehman W (1993) Biochem J294: 63), and of a monoclonal anti-CaD antibody. The results indicated that individual NTF had alternating CaD-rich and CaD-deficient regions. Moreover, we found that the N-termini of all CaD molecules in a given cluster appeared on the same side of an actin filament. Electron microscopic images of crude NTF immunoprecipitated by a polyclonal antibody clearly indicated that the spacing between the CaD clusters is wide enough for myosin heads to interact with actin subunits. Similar clustering of CaD was also observed in plastic embedded tissue sections. These observations raise the possibility that CaD is not acting as a simple on/off switch, but more likely as a modulator, of smooth muscle contraction.

The crystal structure of uncomplexed actin in the ADP state.

The dynamics and polarity of actin filaments are controlled by a conformational change coupled to the hydrolysis of adenosine 5'-triphosphate (ATP) by a mechanism that remains to be elucidated. Actin modified to block polymerization was crystallized in the adenosine 5'-diphosphate (ADP) state, and the structure was solved to 1.54 angstrom resolution. Compared with previous ATP-actin structures from complexes with deoxyribonuclease I, profilin, and gelsolin, monomeric ADP-actin is characterized by a marked conformational change in subdomain 2. The successful crystallization of monomeric actin opens the way to future structure determinations of actin complexes with actin-binding proteins such as myosin.

Phosphorylation of smooth muscle myosin heads regulates the head-induced movement of tropomyosin.

It has been shown that skeletal and smooth muscle myosin heads binding to actin results in the movement of smooth muscle tropomyosin, as revealed by a change in fluorescence resonance energy transfer between a fluorescence donor on tropomyosin and an acceptor on actin (Graceffa, P. (1999) Biochemistry 38, 11984-11992). In this work, tropomyosin movement was similarly monitored as a function of unphosphorylated and phosphorylated smooth muscle myosin double-headed fragment smHMM. In the absence of nucleotide and at low myosin head/actin ratios, only phosphorylated heads induced a change in energy transfer. In the presence of ADP, the effect of head phosphorylation was even more dramatic, in that at all levels of myosin head/actin, phosphorylation was necessary to affect energy transfer. It is proposed that the regulation of tropomyosin position on actin by phosphorylation of myosin heads plays a key role in the regulation of smooth muscle contraction. In contrast, actin-bound caldesmon was not moved by myosin heads at low head/actin ratios, as uncovered by fluorescence resonance energy transfer and disulfide cross-linking between caldesmon and actin. At higher head concentration caldesmon was dissociated from actin, consistent with the multiple binding model for the binding of caldesmon and myosin heads to actin (Chen, Y., and Chalovich, J. M. (1992) Biophys. J. 63, 1063-1070).

Mammal-specific, ERK-dependent, caldesmon phosphorylation in smooth muscle. Quantitation using novel anti-phosphopeptide antibodies.

Extracellular signal-regulated kinases (ERKs) phosphorylate the high molecular mass isoform of the actin-binding protein caldesmon (h-CaD) at two sites (Ser(759) and Ser(789)) during smooth muscle stimulation. To investigate the role of phosphorylation at these sites, antibodies were generated against phosphopeptides analogous to the sequences around Ser(759) and Ser(789). Affinity-purified antibodies were phosho- and sequence-specific. The major site of phosphorylation in h-CaD in porcine carotid arterial muscle strips was at Ser(789); however, the amount of phosphate did not vary appreciably with either KCl or phorbol ester stimulation. Phosphorylation at Ser(759) of h-CaD was almost undetectable (<0.005 mol of phosphate/mol of protein). Moreover, phosphorylation of the low molecular mass isoform of the protein (l-CaD) at the site analogous to Ser(789) was greater in serum-stimulated cultured smooth muscle cells than in serum-starved cells. Serum-stimulated l-CaD phosphorylation was attenuated by the protein kinase inhibitor PD98059. These data 1) identify Ser(789) of h-CaD as the major site of ERK-dependent phosphorylation in carotid arteries; 2) show that the level of phosphorylation at Ser(789) is relatively constant following carotid arterial muscle stimulation, despite an increase in total protein phosphate content; and 3) suggest a functional role for ERK-dependent l-CaD phosphorylation in cell division.

Movement of smooth muscle tropomyosin by myosin heads.

It has been proposed that during the activation of muscle contraction the initial binding of myosin heads to the actin thin filament contributes to switching on the thin filament and that this might involve the movement of actin-bound tropomyosin. The movement of smooth muscle tropomyosin on actin was investigated in this work by measuring the change in distance between specific residues on tropomyosin and actin by fluorescence resonance energy transfer (FRET) as a function of myosin head binding to actin. An energy transfer acceptor was attached to Cys374 of actin and a donor to the tropomyosin heterodimer at either Cys36 of the beta-chain or Cys190 of the alpha-chain. FRET changed for the donor at both positions of tropomyosin upon addition of skeletal or smooth muscle myosin heads, indicating a movement of the whole tropomyosin molecule. The changes in FRET were hyperbolic and saturated at about one head per seven actin subunits, indicating that each head cooperatively affects several tropomyosin molecules, presumably via tropomyosin's end-to-end interaction. ATP, which dissociates myosin from actin, completely reversed the changes in FRET induced by heads, whereas in the presence of ADP the effect of heads was the same as in its absence. The results indicate that myosin with and without ADP, intermediates in the myosin ATPase hydrolytic pathway, are effective regulators of tropomyosin position, which might play a role in the regulation of smooth muscle contraction.

Arrangement of the COOH-terminal and NH2-terminal domains of caldesmon bound to actin.

Smooth muscle caldesmon is a single polypeptide chain with its NH2- and COOH-terminal domains separated by a long alpha-helix. Caldesmon was labeled at either Cys-153 in the NH2 domain or Cys-580 in the COOH domain with a variety of fluorescence probes. Fluorescence intensity, peak position, and polarization of probes on Cys-580 were very sensitive to the binding to actin (with or without tropomyosin), whereas for probes on Cys-153, there was a lack of response, in reconstituted or native actin thin filaments. From fluorescence resonance energy transfer from donor labels on either caldesmon cysteine to acceptor labels on Cys-374 of actin, the distance between the donor and acceptor was estimated to be 27 A for the donor at Cys-580 and 65-80 A for the donor at Cys-153. These findings were the same for caldesmon prepared with or without heat treatment and with striated or smooth muscle actin. These results, together with previous knowledge that COOH-terminal fragments of caldesmon bind to actin whereas NH2-terminal fragments do not, indicate that, while the COOH domain of caldesmon is bound to actin, the NH2 domain is largely dissociated. Fluorescence quenching studies showed that actin binding to caldesmon greatly decreased the accessibility of probes at caldesmon Cys-580 to the quencher, whereas for probes at Cys-153, actin afforded much less, but significant, protection from quenching. Consequently, it appears that, although the NH2 domain is mostly dissociated, it spends some time in the vicinity of actin, through either a weak interaction with actin or collisions with actin and/or because of restricted flexibility which constrains the NH2 domain to be close to the actin filament. Since the NH2 domain of caldesmon binds to the neck region of myosin, a dissociated NH2 domain may account for caldesmon's ability to link myosin and actin filaments.

Strong interaction between caldesmon and calponin.

Caldesmon was labeled at either Cys-153 in the NH2-terminal domain or Cys-580 in the COOH-terminal domain with a 6-acryloyl-2-dimethylaminonaphthalene (acrylodan) fluorescence probe. The addition of smooth muscle calponin to Cys-580-labeled caldesmon resulted in an 18% drop in fluorescence intensity, which titrated with a stoichiometry of 0.9 and a binding constant of 9.5 x 10(7) M-1. For Cys-153-labeled caldesmon, there was no change in fluorescence upon adding calponin. These findings indicate strong binding between calponin and the COOH-domain of caldesmon. The association was sensitive to ionic strength, suggesting that ionic interactions between calponin, a basic protein, and caldesmon, an acidic protein, contribute to the stabilization of the protein complex. That non-muscle acidic calponin interacts with caldesmon with a much reduced association constant of 3.5 x 10(6) M-1 supports such a model. The binding between acidic calponin and caldesmon is strengthened to 1.8 x 10(7) M-1 in the presence of Ca2+, which might bind to acidic residues of the calponin and partially neutralize its negative charge. The strong, specific binding between calponin and caldesmon suggests that this interaction occurs within smooth muscle cells and possibly plays a role in the regulation of contraction.

Cross-linking and fluorescence study of the COOH- and NH2-terminal domains of intact caldesmon bound to actin.

The NH2- and COOH-terminal domains of muscle caldesmon are separated by a long alpha-helical stretch. Cys-580, in the COOH-terminal domain, can be rapidly and efficiently disulfide-cross-linked to Cys-374 of actin by incubation with actin modified with 5,5'-dithiobis(2-nitrobenzoic acid) (Graceffa, P., and Jancso, A. (1991) J. Biol. Chem. 266, 20305-20310). Upon further incubation (+/- tropomyosin), a second cross-link was slowly formed between Cys-153 in the NH2-terminal domain and Cys-374 of another actin monomer. The yield of the second cross-link was relatively insensitive to increasing ionic strength, whereas the caldesmon-actin binding strength decreased considerably, suggesting that the NH2-terminal domain is largely dissociated from actin and becomes slowly cross-linked to it during collisions with the actin filament. In support of these conclusions, the yield of photocross-linking actin to caldesmon specifically labeled with benzophenonemaleimide at Cys-580 was high, but close to zero for caldesmon labeled at Cys-153, and the fluorescence intensity and polarization of tetramethylrhodamine iodoacetamide attached to Cys-580 showed large changes, while there were no changes for the probe at Cys-153 upon binding caldesmon to actin (+/- tropomyosin). These findings are consistent with the knowledge that COOH-terminal fragments of caldesmon bind to actin, whereas NH2-terminal fragments do not. Since the NH2-terminal domain of caldesmon binds to myosin, a dissociated NH2-terminal domain may account for caldesmon's ability to link myosin and actin filaments.

Phytic acid, an iron chelator, attenuates pulmonary inflammation and fibrosis in rats after intratracheal instillation of asbestos.

Reactive oxygen species, especially iron-catalyzed hydroxyl radicals (.OH) are implicated in the pathogenesis of asbestos-induced pulmonary toxicity. We previously demonstrated that phytic acid, an iron chelator, reduces amosite asbestos-induced .OH generation, DNA strand break formation, and injury to cultured pulmonary epithelial cells (268[1995, Am. J. Physiol.(Lung Cell. Mol. Physiol.) 12:L471-480]). To determine whether phytic acid diminishes pulmonary inflammation and fibrosis in rats after a single intratracheal (it) instillation of amosite asbestos, Sprague-Dawley rats were given either saline (1 ml), amosite asbestos (5 mg; 1 ml saline), or amosite treated with phytic acid (500 microM) for 24 hr and then instilled. At various times after asbestos exposure, the rats were euthanized and the lungs were lavaged and examined histologically. A fibrosis score was determined from trichrome-stained specimens. As compared to controls, asbestos elicited a significant pulmonary inflammatory response, as evidence by an increase (approximately 2-fold) in bronchoalveolar lavage (BAL) cell counts at 1 wk and the percentage of BAL neutrophils (PMNs) and giant cells at 2 wk (0.1 vs 6.5% and 1.3 vs 6.1%, respectively; p < 0.05). Asbestos significantly increased the fibrosis score at 2 wk (0 +/- 0 vs 5 +/- 1; p < 0.05). The inflammatory and fibrotic changes were, as expected, observed in the respiratory bronchioles and terminal alveolar duct bifurcations. The increased percentage of BAl PMNs and giant cells persisted at 4 wk, as did the fibrotic changes. Compared to asbestos alone, phytic acid-treated asbestos elicited significantly less BAL PMNs (6.5 vs 1.0%; p < 0.05) and giant cells (6.1 vs 0.2%; p < 0.05) and caused significantly less fibrosis (5 vs 0.8; p < 0.05) 2 wk after exposure. We conclude that asbestos causes pulmonary inflammation and fibrosis in rats after it instillation and that phytic acid reduces these effects. These data support the role of iron-catalyzed free radicals in causing pulmonary toxicity from asbestos in vivo.

Turkey gizzard caldesmon molecular weight and shape.

The molecular weight of chicken gizzard muscle caldesmon has been measured previously by sedimentation equilibrium in the analytical ultracentrifuge and found to be 93 +/- 4 kDa [P. Graceffa, C.-L. A. Wang, and W.F. Stafford (1988) J. Biol. Chem. 263, 14196-14202]. The molecular weight of turkey gizzard caldesmon has been determined by another group to be 75 +/- 2 kDa by the same technique [D.A. Malecik, J. Ausio, C.E. Byles, B. Modrell, and S.R. Anderson (1989) Biochemistry 28, 8227-8233]. We have reevaluated the molecular weight of the turkey protein by sedimentation equilibrium analysis and found a value of 90 +/- 3 kDa, indicating that turkey gizzard caldesmon is a typical muscle caldesmon and does not belong to the class of smaller nonmuscle caldesmons. The two muscle caldesmons do not comigrate during polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate, indicating that they have different amino acid sequences.

Myosin S1 changes the orientation of caldesmon on actin.

Changes in the orientation of caldesmon bound to actin in skeletal ghost myofibrils caused by the binding of myosin subfragment 1 (S1) were measured by fluorescence-detected linear dichroism using fluorescence microscopy. Gizzard caldesmon, labeled with acrylodan at its two Cys residues (CaD*), bound to actin with a probe angle that was unaffected by actin-bound tropomyosin. Irrigation of fibrils with myosin S1 dissociated most of the bound CaD*, but reintroduction of CaD* resulted in its rebinding to actin, without dissociation of S1, with a 7 degrees difference in probe angle. A similar change in probe angle was also observed when a 27-kDa actin-binding C-terminal fragment of caldesmon, labeled with acrylodan at its single Cys 580 (CaD-27*), was used. Introducing MgADP, which bound to S1 in the CaD*-actin-S1 ternary complex in the fibril, reversed the bound CaD* dichroism. These results indicate that (i) myosin heads and caldesmon compete for a common actin binding site; (ii) a ternary complex of CaD*-actin-S1 can be formed with an orientation of CaD* different from that in the CaD*-actin binary complex, and (iii) MgADP, which binds to and reorients myosin S1, affects the orientation of CaD* in the ternary complex. These results are consistent with a two-state binding model of caldesmon for actin in which state 1 involves a site that is competitive with S1 binding and state 2 involves a site that is formed in the presence of bound S1.

Secondary structure and thermal stability of caldesmon and its domains.

Muscle caldesmon is a long, thin protein molecule whose N- and C-terminal regions are separated by a central region which is not present in nonmuscle caldesmon. The three regions appear to be independent structural domains since the alpha-helical content of intact muscle and liver caldesmon is a sum of the alpha-helical contents of the component thrombic fragments over a broad temperature range. Based on circular dichroism spectra of liver and muscle caldesmon and its fragments, together with secondary structure prediction algorithms, it is estimated that the N-domain consists of a string of four to five short-to-intermediate-length alpha-helices; the central domain contains a long continuous alpha-helical stretch; and the C-domain can be divided into two subregions, the N-terminal C1-region, containing a long alpha-helix, and the C-terminal C2-region, containing only random coil. The thermal unfolding of caldesmon takes place gradually without a steep transition and the unfolding is reversible upon cooling, consistent with the known "heat resistance" of caldesmon. This "continuum-of-states" unfolding contrasts with the sharp, cooperative, two-state unfolding characteristic of many proteins. The domains of caldesmon also unfold gradually with the degree of unfolding increasing in the order C-domain < intact molecule < central domain < N-domain, suggesting that the thermal stability decreases in this order.

Disulphide cross-linking of smooth-muscle and non-muscle caldesmon to the C-terminus of actin in reconstituted and native thin filaments.

It was reported that chicken gizzard smooth-muscle caldesmon Cys-580 can be disulphide-cross-linked to the C-terminal pen-ultimate residue (Cys-374) of actin, indicating that these residues are close in the protein complex [Graceffa, P. and Jancso, A. (1991) J. Biol. Chem. 266, 20305-20310]. Since the possibility that the cross-link involves a cysteine residue other than actin Cys-374 was not absolutely excluded, more direct evidence was sought for the identify of the cysteine residues involved in the cross-link. We show here that caldesmon could not be disulphide-cross-linked to actin which had Cys-374 removed by carboxypeptidase A digestion, providing direct support for the participation of actin Cys-374 in the cross-link to caldesmon. In order to assign the caldesmon cysteine residue involved in the cross-link, use was made of caldesmon from porcine stomach muscle, which is shown to contain one cysteine residue close to, or at, position 580, in contrast with chicken gizzard caldesmon, which has an additional cysteine residue at position 153. The porcine stomach caldesmon also formed a disulphide-cross-link to actin, further supporting the original conclusion that Cys-580 of the chicken gizzard caldesmon had been cross-linked to actin. Disulphide-cross-linking with similar yield was also observed in native chicken gizzard muscle thin filaments, indicating that the interaction between actin and the C-terminal domain of caldesmon is the same in native and reconstituted thin filaments. The much smaller non-muscle isoform of caldesmon, from rabbit liver, could be similarly cross-linked to actin, consistent with the sequence similarity between the C-terminal domain of muscle and non-muscle caldesmon. The ability to cross-link caldesmon Cys-580 to actin Cys-374 suggests the possibility that the Cys-580 region of caldesmon and the C-terminus of actin form part of the actin-caldesmon binding interface.

Modification of acidic residues normalizes sodium dodecyl sulfate-polyacrylamide gel electrophoresis of caldesmon and other proteins that migrate anomalously.

Caldesmon migrates as a 140-kDa protein during polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate (SDS), although its true molecular mass is close to 90 kDa. Since caldesmon's high acidic residue content may be responsible for this anomaly, it was reasoned that modification of these residues, with a loss of negative charge, might restore normal electrophoretic migration. Therefore caldesmon was reacted with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide in the presence of excess ethanolamine, which results in negatively charged carboxylates being converted to neutral amides without protein cross-linking. The absence of cross-linking was shown by rotary shadow electron microscopy. In accord with expectations, modified caldesmon migrated as a 94-kDa protein when compared to standards, which were much less affected by modification. The anomalous migration of caldesmon might be due to the repulsion of negatively charged SDS by caldesmon's acidic residues. Low binding of SDS to caldesmon is consistent with the fact that SDS, up to 1%, had little or no effect on the secondary structure of caldesmon, as monitored by circular dichroism. However, other mechanisms can also explain these observations. The abnormal migration of tropomyosin and calsequestrin, both of which have a high percentage of acidic amino acids, was also "normalized" by this treatment. Thus this method might have general application for the electrophoresis of proteins which have a high acidic residue content and migrate anomalously.

Heat-treated smooth muscle tropomyosin.

Gizzard smooth muscle tropomyosin, which is close to 100% gamma beta heterodimer in the native state, was heated to about 100 degrees C, at which temperature the chains are dissociated, followed by reassociation by rapid cooling to 0 degree C. This heat-treated tropomyosin was composed of about 58% heterodimer and 42% of the gamma gamma and beta beta homodimers and had a lower viscosity than that of the native protein, indicating a reduced end-to-end polymerization. Close to 100% heterodimer was regenerated if the heat-treated tropomyosin was subjected to slow cooling from 50 degrees C. However, the viscosity remained low and did not return to the value for untreated tropomyosin, suggesting that the 100 degrees C treatment results in irreversible chemical damage to tropomyosin which affects its end-to-end interaction. Therefore, heat treatment of tropomyosin, a procedure widely used in the preparation of smooth muscle and non-muscle tropomyosins, may result in tropomyosin with a different heterodimer/homodimer distribution and different properties from those of the native protein and should be used with caution.

The role of free radicals in asbestos-induced diseases.

Asbestos exposure causes pulmonary fibrosis and malignant neoplasms by mechanisms that remain uncertain. In this review, we explore the evidence supporting the hypothesis that free radicals and other reactive oxygen species (ROS) are an important mechanism by which asbestos mediates tissue damage. There appears to be at least two principal mechanisms by which asbestos can induce ROS production; one operates in cell-free systems and the other involves mediation by phagocytic cells. Asbestos and other synthetic mineral fibers can generate free radicals in cell-free systems containing atmospheric oxygen. In particular, the hydroxyl radical often appears to be involved, and the iron content of the fibers has an important role in the generation of this reactive radical. However, asbestos also appears to catalyze electron transfer reactions that do not require iron. Iron chelators either inhibit or augment asbestos-catalyzed generation of the hydroxyl radical and/or pathological changes, depending on the chelator and the nature of the asbestos sample used. The second principal mechanism for asbestos-induced ROS generation involves the activation of phagocytic cells. A variety of mineral fibers have been shown to augment the release of reactive oxygen intermediates from phagocytic cells such as neutrophils and alveolar macrophages. The molecular mechanisms involved are unclear but may involve incomplete phagocytosis with subsequent oxidant release, stimulation of the phospholipase C pathway, and/or IgG-fragment receptor activation. Reactive oxygen species are important mediators of asbestos-induced toxicity to a number of pulmonary cells including alveolar macrophages, epithelial cells, mesothelial cells, and endothelial cells. Reactive oxygen species may contribute to the well-known synergistic effects of asbestos and cigarette smoke on the lung, and the reasons for this synergy are discussed. We conclude that there is strong evidence supporting the premise that reactive oxygen species and/or free radicals contribute to asbestos-induced and cigarette smoke/asbestos-induced lung injury and that strategies aimed at reducing the oxidant stress on pulmonary cells may attenuate the deleterious effects of asbestos.

Disulfide cross-linking of caldesmon to actin.

Treatment of a solution of actin and smooth muscle caldesmon with 5,5'-dithiobis(2-nitrobenzoic acid) results in the formation of a disulfide cross-link between the C-terminal penultimate residue Cys-374 of actin and Cys-580 in caldesmon's C-terminal actin-binding region. Therefore, these 2 residues are close in the actin-caldesmon complex. Since myosin also binds to actin in the vicinity of Cys-374 and since caldesmon inhibits actomyosin ATPase activity by the reduction of myosin binding to actin, then the inhibition might be by caldesmon sterically hindering or blocking myosin's interaction with actin. [Ca2+]Calmodulin, which reverses the inhibition of the ATPase activity, decreases the yield of the cross-linked species, suggesting a weakening of the caldesmon-actin interaction in the cross-linked region. It is possible to maximally cross-link one caldesmon molecule/every three actin monomers, in the absence or presence of tropomyosin, clearly ruling out an elongated, end-to-end alignment of caldesmon on the actin filament in vitro, and raising the possibility that the N-terminal part of caldesmon projects out from the filament. Reaction of 5,5'-dithiobis(2-nitrobenzoic acid)-modified actin with caldesmon leads to the same disulfide cross-linked product between actin and caldesmon Cys-580, enabling the specific labeling of the other caldesmon cysteine, residue 153, in the N-terminal part of caldesmon with a spectroscopic probe.

A long helix from the central region of smooth muscle caldesmon.

The central region of smooth muscle caldesmon is predicted to form alpha-helices on the basis of its primary structure. We have isolated a fragment (CT54) that contains this region. The hydrodynamic properties and the electron microscopic images suggest that CT54 is an elongated (35 nm), monomeric molecule. The circular dichroic spectrum yields an overall alpha-helical content of 55-58%. These results are consistent with the model that the middle portion of CT54 forms a long stretch of single-stranded alpha-helix. Such a structure, if it in fact exists, is thought to be stabilized by numerous salt bridges between charged residues at positions i and i + 4. The structural characteristics of this fragment not only represent an unusual protein configuration but also provide information about the functional role of caldesmon in smooth muscle contraction.

Smooth muscle tropomyosin coiled-coil dimers. Subunit composition, assembly, and end-to-end interaction.

Subunits of gizzard smooth muscle tropomyosin, dissociated by guanidinium chloride and reassociated by high salt dialysis, form a 1:1 mixture of the beta beta and gamma gamma homodimers (Graceffa, P. (1989) Biochemistry 28, 1282-1287). The homodimers have now been separated by anion-exchange chromatography and native gel electrophoresis, enabling us to show that the native protein is composed of more than 90% heterodimer. The in vitro equilibrium distribution of heterodimer and homodimers, at close to physiological temperature and ionic conditions, was calculated from thermal unfolding profiles of separated homodimers and heterodimer, as monitored by circular dichroism. The results, for an equal proportion of beta and gamma chains, indicate a predominant formation of heterodimer via chain dissociation and chain exchange, although the proportion of heterodimer was much less than the 90-100% found in the native protein. However, the proportion of heterodimer for actin-bound tropomyosin, determined by analyzing tropomyosin sedimented with actin, was greater than 90%, which may provide a model for assembly in vivo. The end-to-end interactions of the homodimers are about the same but are much less than that of the native heterodimer, as determined by viscometry. The greater end-to-end interaction of heterodimers may lead to stronger binding to actin compared to homodimers and thus would further shift the equilibrium between heterodimer and homodimers toward heterodimer and possibly account for the almost exclusive population of heterodimer in the presence of actin. The greater end-to-end interaction of the heterodimer may also provide a functional advantage for its preferred assembly. This study also shows that the two-step thermal unfolding of the homodimer mixture is due to the formation of heterodimer via an intermediate which is a new type of tropomyosin species which forms a gel in low salt. This tropomyosin is also present in small amounts in native tropomyosin preparations.

Caldesmon from rabbit liver: molecular weight and length by analytical ultracentrifugation.

Although smooth muscle caldesmon migrates as a 140- to 150-kDa protein during sodium dodecyl sulfate-gel electrophoresis, its molecular mass is around 93 kDa as determined by sedimentation equilibrium (P. Graceffa, C-L. A. Wang, and W. F. Stafford, 1988, J. Biol. Chem. 263, 14,196-14,202). Nonmuscle caldesmon migrates during electrophoresis with a molecular mass close to 77 kDa, about half that of the muscle isoform. However, it is controversial whether the molecular weight of nonmuscle caldesmon is the same or much less than that of the muscle protein. Therefore we have now determined the molecular mass of rabbit liver caldesmon by sedimentation equilibrium and found a value of 66 +/- 2 kDa, a value much smaller than that of muscle caldesmon. This new value of the molecular weight, together with a sedimentation coefficient of 2.49 +/- 0.02 S. yields an apparent length of 53 +/- 2 nm and a diameter of 1.7 nm for the liver protein. We previously estimated a length of 74 nm and a diameter of 1.7 nm for the muscle caldesmon. We have also determined the amino acid composition of liver caldesmon and found it to be similar to that of the muscle protein. In conclusion, muscle and nonmuscle caldesmons appear to have similar overall amino acid composition and tertiary structure with the smaller nonmuscle protein having a correspondingly smaller length. The difference in molecular weight between the two caldesmons is consistent with the nonmuscle protein lacking a central peptide of the muscle isoform, as suggested by E. H. Ball, and T. Kovala, (1988, Biochemistry 27, 6093-6098).